1887
banner image
No data available.
Please log in to see this content.
You have no subscription access to this content.
No metrics data to plot.
The attempt to load metrics for this article has failed.
The attempt to plot a graph for these metrics has failed.
Investigation of a glottal related harmonics-to-noise ratio and spectral tilt as indicators of glottal noise in synthesized and human voice signals
Rent:
Rent this article for
USD
10.1121/1.2832651
/content/asa/journal/jasa/123/3/10.1121/1.2832651
http://aip.metastore.ingenta.com/content/asa/journal/jasa/123/3/10.1121/1.2832651

Figures

Image of FIG. 1.
FIG. 1.

Speech harmonics-to-noise ratio (SHNR) for the synthesized vowel /a/ vs for three levels of glottal noise taken from (a) a single spectrum and (b) from averaged spectra

Image of FIG. 2.
FIG. 2.

Glottal harmonics-to-noise ratio (GHNR) vs (a) for three noise levels and (b) noise, shimmer and random and cyclic jitter of the glottal source

Image of FIG. 3.
FIG. 3.

Glottal related harmonics-to-noise ratio (GHNR’) vs (a) for three noise levels and (b) noise, shimmer and random and cyclic jitter of the glottal source

Image of FIG. 4.
FIG. 4.

(a) Ratio of energy below to energy from for the speech signal vs noise, shimmer and random and cyclic jitter of the glottal source. (b) Ratio of energy below to energy from for the speech signal vs noise, shimmer and random and cyclic jitter of the glottal source

Image of FIG. 5.
FIG. 5.

Variation of (a) (speech signal) and (b) (speech signal) with fundamental frequency for three levels of glottal noise of std. dev. 4%, 8%, and 16%

Image of FIG. 6.
FIG. 6.

(Color online) (a) Histogram of HNR (harmonics-to-noise ratio of the speech signal) for the patient/normal data set—not statistically significant at the 5% level (one tailed, two sample, equal variance, student’s t-test). (b) Histogram of GHNR’ (glottal related harmonics-to-noise ratio) for the patient/normal data set—highly statistically significant at the 5% level (one tailed, two sample, equal variance, student’s t-test).

Image of FIG. 7.
FIG. 7.

(Color online) (a) Histogram of (speech signal) for the patient/normal data set—highly statistically significant at the 5% level (one tailed, two sample, equal variance, student’s t-test). (b) Histogram of (speech signal) for the patient/normal data set—not statistically significant at the 5% level (one tailed, two sample, equal variance, student’s t-test).

Image of FIG. 8.
FIG. 8.

Averaged spectra for (a) 8% s.d. glottal noise, (b) 4% s.d. random jitter, (c) 4% cyclic jitter and (d) s.d. 8% random shimmer (window , hop , sampling ) illustrating the reduction in spectral tilt wrt noise and the invariance of spectral tilt wrt temporal and amplitude perturbations.

Image of FIG. 9.
FIG. 9.

Averaged spectra for two “normals,” (a) and (b) and two patients, (c) and (d), of the present study, illustrating the greater spectral tilt for the patient data (window length , hop size and total analysis length ).

Tables

Generic image for table
TABLE I.

Patient listing and details.

Generic image for table
TABLE II.

Ratio of number of harmonics from to the number of harmonics to .

Loading

Article metrics loading...

/content/asa/journal/jasa/123/3/10.1121/1.2832651
2008-03-01
2014-04-23
Loading

Full text loading...

This is a required field
Please enter a valid email address
752b84549af89a08dbdd7fdb8b9568b5 journal.articlezxybnytfddd
Scitation: Investigation of a glottal related harmonics-to-noise ratio and spectral tilt as indicators of glottal noise in synthesized and human voice signals
http://aip.metastore.ingenta.com/content/asa/journal/jasa/123/3/10.1121/1.2832651
10.1121/1.2832651
SEARCH_EXPAND_ITEM